This proposal will evaluate differences in the incidence of diabetes complications and health care costs attributable to alcohol consumption in a large, multiethnic, well-characterized cohort of 65,996 patients with diabetes who receive medical care in an integrated, managed care setting, Kaiser Permanente Northern California. We propose to determine incidence of diabetic complications and all-cause mortality, occurring in groups stratified by alcohol consumption, during up to 11 years follow-up. We will further determine health care costs in each group. Despite the fact that half of U.S. adults consume alcohol, little is known about how this potentially modifiable health behavior affects incidence of complications other than coronary heart disease among people with diabetes. Our study aims are: 1) to evaluate the associations between baseline alcohol consumption (average daily intake, frequency of consumption) and incidence of diabetes complications (myocardial infarction, stroke, congestive heart failure, end-stage renal disease, lower extremity amputation, proliferative retinopathy, acute metabolic events) and all-cause mortality during up to 11 years of follow-up; 2) to determine if ethnicity, gender, or age modify any observed associations between alcohol intake and diabetes complications; 3) to investigate explanatory factors that may link alcohol consumption and diabetes complication incidence rate; 4) to determine the costs of health care (total costs; outpatient, emergency department, and pharmacy costs; hospitalization costs) and 5) to evaluate [unreadable] endogeneity bias, among patients with different levels of alcohol consumption. Using a cohort study design, we will assess these outcomes in 65,996 patients with diabetes who were surveyed in 1994-1997 and have been followed prospectively since then. Our large, ethnically diverse cohort; complete capture of outcomes through automated databases; and extensive characterization of clinical status and other putative confounding variables provides a unique opportunity to study the role of alcohol consumption in the incidence of diabetes complications. Given the incomplete understanding of these epidemiologic relationships, the common use of alcohol, and the increasing public health and economic burden associated with diabetes, these study findings will have important public health significance. [unreadable] [unreadable] [unreadable]